Aberrant signal transduction is a fundamental mechanism underlying malignant growth. The Ras family of signal switch proteins is deregulated in myeloid malignancies by genetic mechanisms that include NRAS and KRAS2 point mutations, the BCR-ABL fusion, PTPN11 mutations, and NF1 inactivation. However, with the notable exception of imatinib mesylate, limited progress has been made toward achieving the goal of developing effective and safe inhibitors of signaling molecules that promote aberrant growth. A major reason for this is that drug discovery is ultimately a biochemical problem that extends beyond our current genetic understanding of cancer biology. A particular bottleneck has been the failure of immortalized cell lines to accurately mimic the behavior of the deregulated signaling networks found in primary cancer cells. In previous studies supported by this award, we showed that the NF1 gene, which encodes a GTPase activating protein (GAP) for Ras, functions as a tumor suppressor gene in immature myeloid cells by negatively regulating Ras signaling. We also exploited a "first generation" mouse model of myeloproliferative disease (MPD) that results from transplanting homozygous Nf1 mutant liver cells into irradiated recipient mice to demonstrate that GM-CSF plays a central role in the abnormal proliferation of Nf1-deficient hematopoietic cells, and to perform preclinical studies of an inhibitor of the Ras processing enzyme farnesyltransferase. In the current period of support, we developed strains of mice that accurately model the genetic and biochemical consequences of hyperactive Ras in primary hematopoietic cells by using the Mx1- Cre strain to ablate a conditional mutant Nf1 allele or to activate oncogenic Kras expression from its endogenous promoter. The goals of this renewal application are to characterize how leukemia-associated genetic lesions deregulate Ras-regulated signaling networks, to identify targets for therapeutic intervention, to use these tractable new models as platforms for testing novel agents, and to identify genes and pathways that cooperate with hyperactive Ras in myeloid leukemogenesis. We propose three specific aims: (1) To characterize and compare the biochemical consequences of oncogenic KrasG12D expression and Nf1 inactivation by interrogating a network of signaling molecules in primary bone marrow cells and in defined subsets of stem/progenitor cells. (2) To investigate the contributions of specific Ras effectors to aberrant growth by expressing mutant proteins in primary hematopoietic cells and by generating and analyzing novel strains of "knock in" mice. (3) To utilize retroviral insertional mutagenesis as a general strategy for uncovering genes that cooperate with oncogenic Kras to induce progression from MPD to AML, and to test how these mutations modulate therapeutic responses in vitro and in vivo.